This is not really applicable to the Pathfinder RPG, at least not as far as the first edition is concerned, because it is far too radical, but I would still like to get some comments on this.
Suppose that instead of bonuses being a flat number, they would be rolled as extra dice. There are two basic ways of doing this: the average method and the maximum method.
Average Method: (two options, lower or higher bonus)
Original Bonus:...New Bonus (Lower):...New Bonus (Higher):
+1................+1...................+d2
+2................+d2..................+d4
+3................+d4..................+d6
+4................+d6..................+d8
+5................+d8..................+d10
Etcetera..........Etcetera.............Etcetera
If chosing the average method, I would chose the lower bonuses (middle column) and yes, I know that +2 could be perfectly averaged by d3, but I am making it a d2 or d4 for the sake of consistency with the other numbers.
The advantages of using the average method are that static numbers don't need to be recalculated to get the same average results.
Maximum Method:
Original Bonus:...New Bonus:
+1................+1
+2................+d2
+3................+d3
+4................+d4
+5................+d5
Etcetera..........Etcetera
The disadvantage of using the maximum method is that in order to get the same average results in terms of difficulty, static numbers would need to be recalculated (all deviadions from the base number of 10(.5)would have to be halved).
Both the average method and the maximum method suffer from the same problem that the types of dice we have are limited - we don't have a commonly available d7 for example, nor can we easily simulate it. Computers can deal with this, of course, but reliance on computers is not desirable. As such, it may be necessary to stop at a certain number and start adding dice anew. A good number to stop at would be the original +5 bonus, since it is a nice and neat quarter of 20, the maximum attainable with an unmodified d20 roll, but any other number for which we have dice available would be reasonable (stopping at d4, d6, d8, d10, d12...). Beyond the stop point, new dice would be added, so starting from +1 again, while keeping the previously gained dice too.
The purpose of this system would be to ensure that even when the spread of bonuses between characters is very high, there is still a chance of failure for the characters with higher bonuses and a chance of success for those with lower bonuses. So somebody (e.g. Fighter) with a Reflex saving throw bonus of +5 would instead add an extra d8 to his roll, whereas another character (e.g. Rogue) with a Reflex saving throw of +13 would instead add an extra 2d8 and an extra d4 to the roll (this example assumes the use of the average method (low bonus) and assumes that the stopping point for increasing dice size is set at the d8).
|
GURPS simulates this by having every +4 turn into +1d6 and every +7 turn into +2d6.
Converting this over to D&D, a longsword attack that did 1d8+12 would do 1d8 +3d6 +1. (The first +7 turns into +2d6, the next +4 turns into another +1d6, and there's a +1 leftover.) Using this system, no damaging effect will ever have more than +3 damage, as +4 converts into +1d6.
The advantage is potential bigger damage and potential smaller damage, for narrative situations where the DM wants to say that an NPC survived a hit that they shouldn't have (oh, he took minimum damage, which would have been 13 hp, but under this swingier system is 5 hp) or, alternately, wants to have an NPC die like a chump (yes, he had 12 hp, which means that 1d8+12 couldn't actually drop him to -10, but using the swingier system, he could take up to 27 hp damage from that hit!).
A disadvantage is the swinginess resulting in spectacularly more damage than the DM anticipated (or less damage from the PCs) creating a greater chance of an accidental player kill. (Only compounded by the effects of criticals, obviously.)
I'm not sure that D&D needs this sort of thing, and I prefer the reliability of numbers like 1d8+12, but it's certainly an option.
GURPS simulates this by having every +4 turn into +1d6 and every +7 turn into +2d6.
Converting this over to D&D, a longsword attack that did 1d8+12 would do 1d8 +3d6 +1. (The first +7 turns into +2d6, the next +4 turns into another +1d6, and there's a +1 leftover.) Using this system, no damaging effect will ever have more than +3 damage, as +4 converts into +1d6.
The advantage is potential bigger damage and potential smaller damage, for narrative situations where the DM wants to say that an NPC survived a hit that they shouldn't have (oh, he took minimum damage, which would have been 13 hp, but under this swingier system is 5 hp) or, alternately, wants to have an NPC die like a chump (yes, he had 12 hp, which means that 1d8+12 couldn't actually drop him to -10, but using the swingier system, he could take up to 27 hp damage from that hit!).
A disadvantage is the swinginess resulting in spectacularly more damage than the DM anticipated (or less damage from the PCs) creating a greater chance of an accidental player kill. (Only compounded by the effects of criticals, obviously.)
I'm not sure that D&D needs this sort of thing, and I prefer the reliability of numbers like 1d8+12, but it's certainly an option.
Interesting - thanks for explaining how GURPS does this.
I actually meant something a bit different though. Damage would stay the same as it is now under this system. It is the bonuses to d20 rolls that would be affected.
So the affected rolls would be:
Attack Rolls
Saving Throw Rolls
Skill Checks
Ability Checks
Caster Level Checks
Etcetera
Bonuses to these rolls would be converted to dice as I described above. This would make the sweet spot less important and extend its range by a significant margin, yet still allow progressions with increasing differences between the classes.
I'd rather not Pathfinder change to a dice pool system. Stick to bonuses for the most part!
I don't like this, sorry
the "average method" is an incorrect application of a bell curve
the "maximum method" is no different than using larger dice than a d20 and adding +1
the current system scales by injecting a minimum result in the equation
this represents overcoming the lesser problems in an adventurer's life
I'd rather not Pathfinder change to a dice pool system. Stick to bonuses for the most part!
This change would be too radical for Pathfinder (at least the first version), so I agree. I just want comments about it in general.
I don't like this, sorry
No problem - no need to appologize - we all like different things.
the "average method" is an incorrect application of a bell curve
I am not sure what you mean by this. Could you explain? The "average method" is called that because it yields the same (or almost the same) average results as the character would obtain by using the old bonus, but the character can also fail or succeed in situations where it would have been impossible with the legacy system.
the current system scales by injecting a minimum result in the equation
this represents overcoming the lesser problems in an adventurer's life
This is correct, but it leads to eventually bypassing the so called 'sweet spot' at which point the spread in bonuses between different characters becomes so large that either some succeed automatically or some fail automatically, making it difficult to provided challenges for the party as a whole. That's precisely the reason I am gathering comments on this system, which helps to avert this situation and prolong the 'sweet spot' yet allow differential bonus progressions. I am not saying it is necessarily a good system, but that is the reason for me creating it and gathering comments. I think some people will just dislike the system because of tradition and aesthetics, which is a perfectly valid reason to dislike a system, but there may also be mechanical reasons why people dislike the system.
|
I'd rather not Pathfinder change to a dice pool system. Stick to bonuses for the most part!This change would be too radical for Pathfinder (at least the first version), so I agree. I just want comments about it in general.
Thanks for sharing, Roman! It's an interesting idea, but I agree the logistical difficulties exist.
Another possibility is that rather than adding dice onto the rolls, bonuses increase more slowly, and we add additional dice to rolls and let the player choose the higher roll.
One possible implementation of the idea:
+1 BaB characters now = 1/2 BaB, with an extra 1d20 to roll each 5 levels
+3/4 BaB characters now = 1/2 BaB with an extra 1d20 each 10 levels
+1/2 BaB character now = 1/4 BaB with an extra 1d20 each 10 levels
Thus a 16th level fighter has BaB +8, with 4d20 rolled each attack - and only the most desired roll kept.
A 16th level rogue would have BaB +8, but would only roll 2d20 each attack - and keep the highest.
And a 16th level sorceror would have BaB +4, and would roll 2d20 and keep the highest.
This means the fighter will only fumble extremely rarely, and miss more rarely than he does now as well - but will keep the bonuses within a much smaller range of numbers. He will crit much more often as well. Criticals become more of a function of skill instead of luck. This also prevents the frustrating flat 5% fumble chance for martial characters - meaning that in a toe to toe full attack only combat under the current system, your fighter will typically fumble every 24 seconds, while a wizard foolishly engaged in the same sort of toe to toe combat will only fumble every 48 seconds. I've always found that an intensely frustrating artifact of the rules.
agree with the "sweet spot" problem, auto successes do appear too soon
maybe progession should be slower there or taper off - dunno
have seen a lot of 1's rolled for the odd trivial matter by high level chars though rotfl :D
seems to hurt player pride a lot more than the character ;)
what I meant by bell curve use, is there is no average when using 1 die
there is an equal chance of rolling any number
the more times rolled, the more the results flat line
trends occur which can give a person the illusion otherwise
especially if prone to gambling addiction "if this happens, or I do that, then the die tends to roll x"
on the other hand, two or more dice produce an average. for instance, 2d6 average is 7
there are more combinations that equal 7 than any other, with 2 and 12 being unique combinations
that produces a natural bell curve
was just in a fussy mood when I read it is all lol ;)
What about poor schlubbs like me who have terrible luck rolling dice? This would be as welcome as a red-hot poker in the eye... :~(
|
on the other hand, two or more dice produce an average. for instance, 2d6 average is 7
there are more combinations that equal 7 than any other, with 2 and 12 being unique combinations
that produces a natural bell curve
In statistics, this phenomenon is called the Central Limit Theorem.
Just FYI.
-Skeld
Just wanted to say that I *love* this concept, just hate the implementation.
I'm not one to normally say "we roll too many dice!", but when you hit that ranger firing an anarchic, holy longbow with flaming arrows at a target she's got Bane of Enemies for, five times in a round...
Well, suffice to say I wanted to avoid "flat" averages enough that I developed a chart, where a single die roll (of any size you wanted) tied into the results of the curve of results.
So for a 10d6 roll, you wouldn't take the 35 average, or roll 1d6 x 10 (shudder). Instead you'd roll a d20, or d6, and get something like this (I don't remember the exact values atm):
10d6 on d6:
1 - 23
2 - 29
3 - 33
4 - 37
5 - 41
6 - 47
Worked well, but not exactly easy to put together :)
|
|
Here's a quick graph of what the percentage change of getting a certain d20 result would be when rolling 1, 2, 3 or 4 d20s and taking only the maximum value:
It was taking too much time to calculate the values, so I stopped at 4d20, even though the proposal above went to 5d20.
Any which way, something like that would give martial characters a much better chance for success without stacking the differences between bonuses for classes too high.
Another possibility is that rather than adding dice onto the rolls, bonuses increase more slowly, and we add additional dice to rolls and let the player choose the higher roll.One possible implementation of the idea:
+1 BaB characters now = 1/2 BaB, with an extra 1d20 to roll each 5 levels
+3/4 BaB characters now = 1/2 BaB with an extra 1d20 each 10 levels
+1/2 BaB character now = 1/4 BaB with an extra 1d20 each 10 levelsThus a 16th level fighter has BaB +8, with 4d20 rolled each attack - and only the most desired roll kept.
A 16th level rogue would have BaB +8, but would only roll 2d20 each attack - and keep the highest.
And a 16th level sorceror would have BaB +4, and would roll 2d20 and keep the highest.
This means the fighter will only fumble extremely rarely, and miss more rarely than he does now as well - but will keep the bonuses within a much smaller range of numbers. He will crit much more often as well. Criticals become more of a function of skill instead of luck. This also prevents the frustrating flat 5% fumble chance for martial characters - meaning that in a toe to toe full attack only combat under the current system, your fighter will typically fumble every 24 seconds, while a wizard foolishly engaged in the same sort of toe to toe combat will only fumble every 48 seconds. I've always found that an intensely frustrating artifact of the rules.
This is also an interesting idea and leads to a somewhat different dynamics, because multiple rolls with the ability to choose the highest increase the chance of getting a higher number without increasing the maximum that can be reached. This could work, but would necessitate the recalculation of ACs and DCs to take this into account.
what I meant by bell curve use, is there is no average when using 1 die
there is an equal chance of rolling any number
the more times rolled, the more the results flat line
trends occur which can give a person the illusion otherwise
especially if prone to gambling addiction "if this happens, or I do that, then the die tends to roll x"on the other hand, two or more dice produce an average. for instance, 2d6 average is 7
there are more combinations that equal 7 than any other, with 2 and 12 being unique combinations
that produces a natural bell curve
OK, I think you are confusing distribution with average/mean. A d6 die does have a mean/expected value and it is 3.5. The distribution of a single die is uniform, yes, but the mean value still exists. If we have more than one die than we start to get an approximation to the normal distribution and thus the bell curve, but averages/means/expected results exist even in distributions without bell curves.
Just wanted to say that I *love* this concept, just hate the implementation.
I'm not one to normally say "we roll too many dice!", but when you hit that ranger firing an anarchic, holy longbow with flaming arrows at a target she's got Bane of Enemies for, five times in a round...
This is a valid mechanical concern - thanks for bringing it up.
Well, suffice to say I wanted to avoid "flat" averages enough that I developed a chart, where a single die roll (of any size you wanted) tied into the results of the curve of results.
So for a 10d6 roll, you wouldn't take the 35 average, or roll 1d6 x 10 (shudder). Instead you'd roll a d20, or d6, and get something like this (I don't remember the exact values atm):
10d6 on d6:
1 - 23
2 - 29
3 - 33
4 - 37
5 - 41
6 - 47
Interesting - did you basically run the probabilities for all dice combinations and mapped them onto a d6 die? I can see why you would want to do this, but it would take a truly massive table to do that!
OK, so rolling too many dice is odious and we should probably stick to d20s. How about something like this to deal with the sweetspot issue:
The resolution mechanism is the standard d20 roll and the bonus to the roll is added normally with one caveat. The maximum bonus that can be added to the roll cannot exceed the natural roll itself. This means that if a character has a bonus of +7 to a Fortitude saving throw and he rolls a 15, he adds the full +7 for a total result of 22. Should he, however, only roll 4, he can only add +4 to the roll for a total result of 8.
Of course, this cannot go on forever, since bonuses above +20 are higher than the maximum number attainable on a d20 and they need to remain meaningful. There are two reasonable possibilities as to what could be done with bonuses higher than 20 (other than not adding them at all). Either they could simply be added regardless of the what is rolled, or they could again be capped from the beginning (if the latter model were chosen, perhaps the recap point should be lower than 20, maybe 10 or so).
This system would ensure that even a very high spread between the bonuses of characters would not prevent them being at least somewhat challenged by the same threats.
I've long had damage bonuses add to the size of the die, e.g., d8+3 becomes d11. But when the whole system is built on stacking bonuses that doesn't work as well.
I'd rather not Pathfinder change to a dice pool system. Stick to bonuses for the most part!This change would be too radical for Pathfinder (at least the first version), so I agree. I just want comments about it in general.
That was my comment in general. I'm not against dice pool systems (I like the WoD system as well as L5R, which has a really cool system) in general. I just don't want them in my D&D/PF.
What about poor schlubbs like me who have terrible luck rolling dice? This would be as welcome as a red-hot poker in the eye... :~(
Yeah. Though I have to say that on Sunday, we (especially our party's fighter) was saved because the level 9 priestess managed to deal 15 (!) points of damage with her flame strike. (For the record: Average is 31.5). Sometimes, even GMs have bad luck rolling dice.
All I can say is that for certain members of my gaming group, we would need dozens of dice for every attack. Just think of what happens when a frenzied beserker Power Attacks for full on a Leap Attack! (before you call me on referencing non-core material, just know that people will find ways to break Pathfinder as well... especially if supplements are ever released)
OK, I think you are confusing distribution with average/mean. A d6 die does have a mean/expected value and it is 3.5. The distribution of a single die is uniform, yes, but the mean value still exists. If we have more than one die than we start to get an approximation to the normal distribution and thus the bell curve, but averages/means/expected results exist even in distributions without bell curves.what I meant by bell curve use, is there is no average when using 1 die
there is an equal chance of rolling any number
the more times rolled, the more the results flat line
trends occur which can give a person the illusion otherwise
especially if prone to gambling addiction "if this happens, or I do that, then the die tends to roll x"on the other hand, two or more dice produce an average. for instance, 2d6 average is 7
there are more combinations that equal 7 than any other, with 2 and 12 being unique combinations
that produces a natural bell curve
incorrect, I was not getting anything confused, I just replied in layman terms
like I said, I was having a "fussy" moment, which you seem to be having here now rotfl :Pwe just disagree, that's all silly - relax :D
I come from a background that deals in absolutes not expected values
no reason for us to lock heads over it, let our professors duke that out ;)
Interesting - did you basically run the probabilities for all dice combinations and mapped them onto a d6 die? I can see why you would want to do this, but it would take a truly massive table to do that!Well, suffice to say I wanted to avoid "flat" averages enough that I developed a chart, where a single die roll (of any size you wanted) tied into the results of the curve of results.
So for a 10d6 roll, you wouldn't take the 35 average, or roll 1d6 x 10 (shudder). Instead you'd roll a d20, or d6, and get something like this (I don't remember the exact values atm):
10d6 on d6:
1 - 23
2 - 29
3 - 33
4 - 37
5 - 41
6 - 47
Yep, that's exactly... wait no. No I didn't run it for *all* possibilities, I did 50K runs of whatever scenario I was doing, and mapped that out - averaging the sectors (whether d6 or d20). I still have the excel so I can generate more if I really want to.
Found my chart, follow the Link.
This is not really applicable to the Pathfinder RPG, at least not as far as the first edition is concerned, because it is far too radical, but I would still like to get some comments on this.
Suppose that instead of bonuses being a flat number, they would be rolled as extra dice. There are two basic ways of doing this: the average method and the maximum method.
Or maybe something imbetween that works better with die sizes we already know and love. Roll the die size nearest half the total bonus, and add enough to equal the bonus on a maximum roll.
Static . Dynamic
Bonus . Bonus
+1 . . . . . +1
+2 . . . . . +2
+3 . . . . . +3
+4 . . . . d2+2
+5 . . . . d2+3
+6 . . . . d3+3
+7 . . . . d3+4
+8 . . . . d4+4
+9 . . . . d4+5
+10 . . . d6+4
+11 . . . d6+5
+12 . . . d6+6
+13 . . . d6+7
+14 . . . d6+8
+15 . . . d8+7
+16 . . . d8+8
Interesting - did you basically run the probabilities for all dice combinations and mapped them onto a d6 die? I can see why you would want to do this, but it would take a truly massive table to do that!Well, suffice to say I wanted to avoid "flat" averages enough that I developed a chart, where a single die roll (of any size you wanted) tied into the results of the curve of results.
So for a 10d6 roll, you wouldn't take the 35 average, or roll 1d6 x 10 (shudder). Instead you'd roll a d20, or d6, and get something like this (I don't remember the exact values atm):
10d6 on d6:
1 - 23
2 - 29
3 - 33
4 - 37
5 - 41
6 - 47
Yep, that's exactly... wait no. No I didn't run it for *all* possibilities, I did 50K runs of whatever scenario I was doing, and mapped that out - averaging the sectors (whether d6 or d20). I still have the excel so I can generate more if I really want to.
Found my chart, follow the Link.
Thanks for the link! That could be pretty useful for speeding up play.
This is not really applicable to the Pathfinder RPG, at least not as far as the first edition is concerned, because it is far too radical, but I would still like to get some comments on this.
Suppose that instead of bonuses being a flat number, they would be rolled as extra dice. There are two basic ways of doing this: the average method and the maximum method.
Or maybe something imbetween that works better with die sizes we already know and love. Roll the die size nearest half the total bonus, and add enough to equal the bonus on a maximum roll.
Static . Dynamic
Bonus . Bonus
+1 . . . . . +1
+2 . . . . . +2
+3 . . . . . +3
+4 . . . . d2+2
+5 . . . . d2+3
+6 . . . . d3+3
+7 . . . . d3+4
+8 . . . . d4+4
+9 . . . . d4+5
+10 . . . d6+4
+11 . . . d6+5
+12 . . . d6+6
+13 . . . d6+7
+14 . . . d6+8
+15 . . . d8+7
+16 . . . d8+8
Not bad - an intermediate system that keeps half the bonus and converts half to a die - it certainly involves less rolling than a full conversion to dice.
The problem I see in this suggestions is related to bonus damage to critical hits.
page 131:
"Multiplying Damage: Sometimes you multiply damage by some factor, such as on a critical hit. Roll the damage (with all modifiers) multiple times and total the results. Note: When you multiply damage more than once, each multiplier works off the original, unmultiplied damage.
Exception: Extra damage dice over and above a weapon’s normal damage are never multiplied."
So, either we change also the rules for critical hits, or this rule would hamper critical hit damage immensely.
This is a problem the Paladin is currently facing with the alternate Smite Evil feature (flat bonus equal to Paladin level vs Evil creatures; the bonus increases to 1d6 extra damage per 2 Paladin levels vs Evil Outsiders).
A Paladin facing a Fiend is generally best suited... unless he scores a Critical Hit (the extra d6 are NOT multiplied).
Worse yet, a Spirited Charge with a Lance (damage x3) is best suited vs a generic Evil creature, since the flat bonus is multiplied x3 as well; against an Evil Outisider, no luck...
10th-level Paladin, STR 10, Smite Evil Spirited Charge vs. Evil creature: 3d8+30 (triple damage for 1d8+10), critical 5d8+50*
average: 43 (critical 72)
10th-level Paladin, STR 10, Smite Evil Spirited Charge vs. Evil Outsider: 3d8 (triple damage from 1d8...) plus 5d6, critical 5d8* plus 5d6
average: 31 (critical 40)
*by SRD rules:
"Sometimes a rule makes you multiply a number or a die roll. As long as you’re applying a single multiplier, multiply the number normally. When two or more multipliers apply to any abstract value (such as a modifier or a die roll), however, combine them into a single multiple, with each extra multiple adding 1 less than its value to the first multiple. Thus, a double (×2) and a double (×2) applied to the same number results in a triple (×3, because 2 + 1 = 3)."
in this case triple (x3) from Spirited Charge and triple (x3) from Critical = (x5)
Just my 2c.
RPG Superstar 2010 Top 16
|
Another possibility is that rather than adding dice onto the rolls, bonuses increase more slowly, and we add additional dice to rolls and let the player choose the higher roll.One possible implementation of the idea:
+1 BaB characters now = 1/2 BaB, with an extra 1d20 to roll each 5 levels
+3/4 BaB characters now = 1/2 BaB with an extra 1d20 each 10 levels
+1/2 BaB character now = 1/4 BaB with an extra 1d20 each 10 levelsThis means the fighter will only fumble extremely rarely, and miss more rarely than he does now as well - but will keep the bonuses within a much smaller range of numbers. He will crit much more often as well. Criticals become more of a function of skill instead of luck.
Jess, how would you implement a Full Attack suite? Would you use the second, and possibly third highest dice as theiterative attacks?
I have to say, few players that I have ever gamed with were able to rattle off addition like they were Rain Man.
Give them d20+17 and they take a short while ticking off numbers, some take a shorter while than others, some take a longer while - sometimes they even counted on their fingers.
But give them a d20 +3 +5 +4 +3 +2 (representing some extra dice that were rolled) and this will just slow them down more.
I venture to say that even people who are good at doing addition in their head will always add a pair of numbers faster than they will add a series of 5 or 6 or more numbers.
Why would we want to do this and slow down combat?
Surely we can come up with a better way to reduce auto-success, or simply learn to accept that high level characters are *supposed* to be superheroic and consequently, they are *supposed* to make difficult tasks look easy, or even automatic.
The problem I see in this suggestions is related to bonus damage to critical hits.
page 131:
"Multiplying Damage: Sometimes you multiply damage by some factor, such as on a critical hit. Roll the damage (with all modifiers) multiple times and total the results. Note: When you multiply damage more than once, each multiplier works off the original, unmultiplied damage.
Exception: Extra damage dice over and above a weapon’s normal damage are never multiplied."So, either we change also the rules for critical hits, or this rule would hamper critical hit damage immensely.
Even though dice are rolled, this is still counted as a numerical bonus rather than bonus dice. The value of the number just happens to vary to keep huge bonuses in check.